Equation of State Compositional Modeling on Vector Processors

Abstract

Equation of State Compositional Modeling on Vector Processors L.C. Young L.C. Young Reservoir Simulation Research Corp. Search for other works by this author on: This Site Google Scholar Paper presented at the SPE Symposium on Reservoir Simulation, San Antonio, Texas, February 1987. Paper Number: SPE-16023-MS https://doi.org/10.2118/16023-MS Published: February 01 1987 Cite View This Citation Add to Citation Manager Share Icon Share Twitter LinkedIn Get Permissions Search Site Citation Young, L.C. "Equation of State Compositional Modeling on Vector Processors." Paper presented at the SPE Symposium on Reservoir Simulation, San Antonio, Texas, February 1987. doi: https://doi.org/10.2118/16023-MS Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex Search nav search search input Search input auto suggest search filter All ContentAll ProceedingsSociety of Petroleum Engineers (SPE)SPE Reservoir Simulation Conference Search Advanced Search SPE MemberAbstractPreviously, large simulation problems have not been feasible with equation of state (EOS) compositional models. This paper describes an approach for efficient EOS calculations. Tests of a simulator based on the approach, show that large problems can now be solved. For example, a problem composed of 9-components, 20,000 grid points, and 1,000 time steps can be solved in approximately 40 minutes of CPU time on a Cray X-MP.IntroductionIn the past, applications using equation of state (EOS) compositional models have been restricted due to large computation times. Engineers have often been faced with a choice either to use a black oil model and neglect important compositional effects or to use a compositional model with grids too coarse to resolve important flow effects. For example, the largest reported compositional study consisted of approximately 8,000 grid points, 9 hydrocarbon phase components and required over 8 hours of computing time on an IBM 3033 computer. Such large computing requirements has prohibited the use of EOS models on larger problems. This paper describes a simulator formulated to overcome this limitation.Before this project was initiated, the following goals and objectives were outlined:To develop a simulator which will run large problems on a vector computer, atleast 20 times faster than the scalar time reported in reference 1.To develop a formulation which also executes efficiently on conventional scalar machines.To use a general formulation which can accommodate other fluid property correlations, e.g. black oil fluids.To achieve high portability by using ANSI standard FORTRAN 77 as much as possible.For a typical compositional simulation, more than 50% of the computation time is spent in the EOS calculations. Obviously, high performance cannot be achieved unless these calculations are executed efficiently. With most previous simulators, these calculations either will not vectorize or they will vectorize with short vector lengths, e.g. equal to the number of components or the number of contiguous grid points with a given "fluid type". The term "fluid type" classifies points by the calculations required. For example, the sequence of calculations performed for a single hydrocarbon phase is different from that performed for two hydrocarbon phases.Efficient calculations on vector machines need vector lengths equal to the total number of points with a given fluid type. The need for long vector lengths dictates that the points be ordered by fluid type rather than by grid point number. The method described in the body of the paper maintains two orderings of the grid points, one based on the computational grid and one based on fluid type. To minimize the work associated with reordering the data, the major fluid property arrays always remain ordered by fluid type. These arrays are rearranged only when the fluid at some point changes from one fluid type to another.This paper is divided into several parts.P. 313^ Keywords: fluid type, upstream oil & gas, vector length, grid point, iteration, fluid modeling, equation of state, time step, reservoir simulation, calculation Subjects: Fluid Characterization, Reservoir Simulation, Fluid modeling, equations of state This content is only available via PDF. 1987. Society of Petroleum Engineers You can access this article if you purchase or spend a download.